-
Validation of cloud and aerosol classification results based on three years
MAX-DOAS observations in Wuxi (China) using independent data sets Yang Wang
1( Email: [email protected] ), Thomas Wagner
1, Pinhua Xie2, Steffen Beirle1, Steffen Dörner1, Julia Remmers
1, Ang Li
2
1) Max-Planck institute for Chemistry, satellite group, Mainz, Germany
2) Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei, China
Abstract
Multi-Axis-Differential Optical Absorption Spectroscopy (MAX-DOAS) observations of trace
gases can be strongly influenced by clouds and aerosols. Thus it is important to identify clouds
and characterise their properties. In the former work (Wagner et al. 2013) we found the colour
index, radiance and O4 absorption from MAX-DOAS measurements are sensitive to the prop-
erties of cloud and aerosol and built a sophisticated classification scheme. In this work we fur-
ther improved the identification of clouds and aerosol for each elevation sequence of MAX-
DOAS based on three years of measurements (2011 to 2013) in Wuxi, China (31.57°N,
120.31°E). The cloud classification results were verified by comparing with other cloud or
aerosol data sets such as the aerosol optical depth (AOD) from the AERONET Taihu monitor-
ing site (31.42° N, 120.22° E), MODIS Level 2 cloud products and cloud parameters in level
2b productions of OMI and GOME-2 from TEMIS. We find good agreement with the MAX-
DOAS cloud classification using statistical analyses. Based on the results of MAX-DOAS
cloud classification, we investigate the validation of tropospheric NO2 VCD from OMI with
the clouds tropospheric NO2 VCD from MAX-DOAS. The flags of sky conditions provide
more information on the validation of satellite productions by MAX-DOAS observations.
Cloud classification scheme of MAX-DOAS
Motivation ● The effects of sky conditions (clear sky, continuous clouds, broken clouds and high aerosol
load) on MAX-DOAS measurements are quite different.
● It is necessary for processing or interpolation of MAX-DOAS data to flag the sky conditions
for each measurement.
● It is helpful to precisely validate the tropospheric vertical column density from satellite
products in different sky conditions.
We want to build a convincing scheme to derive cloud and aerosol information (sky con-
ditions) from the individual MAX-DOAS observations
Quantity for cloud classification abbrevi-
ation
thresh
old
Normalized CI (340nm/420nm) for zenith
view n
zCI 0.9
Normalized temporal smoothness indicator of
CI for zenith view n
zTSI 8×10
-8 s-2
Sum of normalized temporal smoothness indi-
cator of CI for non-zenith view n
LTSI 2.5×
10-7
spread of the CI SpreadCI 0.3
Normalized O4 AMF 4nO
AMF 0.8
Normalized radiance at 380 nm Radian-
cen 0.9
spread of the O4 dAMF SpreadO
4 0.4
The classification scheme is based on seven quantities deduced from the radiance, CI and O4
absorption observed by MAX-DOAS. The seven quantities are shown in the table 1. Based
on the quantities, we built this scheme in figure 1. Its left column show the determinations
and the right blue column show the deduced sky conditions. We identify the sky conditions
by comparing the quantities from individual MAX-DOAS observations with their corre-
sponding reference values in clear sky. The scheme include two kinds of classifications. One is
primary classifications indicated by the black arrows and another one is additional classifica-
tions indicated by the blue arrows. Note that The sky condition for each MAX-DOAS meas-
urement should belong to one primary classification. In addition to the primary classification
additional secondary classifications can be assigned.
Validation of the scheme based on MAX-DOAS Measure-
ments in Wuxi, China
MAX-DOAS: The measurements are operated from 1
May 2011 to 29 Nov 2013 with 5 elevation angles (5°,
10°, 20°, 30° and 90°), azimuth angle is 0 °(north)
Other techniques for comparisons with MAX-DOAS:
1. Visibilitymeter: the visibility at 550 nm
2. Aeronet: The AOD is available from sun photometer
operated at Taihu about 18 km west-south away from
the Wuxi MAX-DOAS site. 3. effective cloud fraction form OMI, GOME-2 4. geometrical CF from MODIS 5. cloud optical thickness from MODIS
Validation of the scheme based on MAX-DOAS Meas-
urements in Wuxi, China
0.0 0.5 1.0 1.5 2.0 2.5 3.00%
50%
100%
150%
200%
p
erce
nt
of
sky
co
nd
itio
ns
(a)
0.0 0.5 1.0 1.5 2.0 2.5 3.00
1000
2000
3000
4000
5000
AOD at 340 nm
nu
mb
er o
f sk
y c
on
dit
ion
s
0 20 40 60 800
50100150200 clear with low aerosol strong visibal absorption clear with high aerosol
zenith cloud holes offzenith cloud holes broken clouds continuous clouds
exceptional continuous clouds fog "or" thick clouds "and" thick clouds
0 10 20 300%
50%
100%
150%
200%
visibility at 550 nm
num
ber
s of
sky c
ondit
ions
(b)
0 10 20 300
2000
4000
6000
8000
10000
12000
per
cent
of
sky c
ondit
ions
0 20 40 60 800%
50%
100%
150%
200%
250%
effective CF from OMI
nu
mb
er
of
sky c
on
ditio
ns
pe
rce
nt
of
sky c
on
ditio
ns
(c)
0 20 40 60 800
50
100
150
200
0 20 40 60 800%
50%
100%
150%
200%
250%
300%
0 20 40 60 800
50
100
150
200
(d)
effective CF from GOME-2
nu
mb
er
of
sky c
on
ditio
ns
pe
rce
nt
of
sky c
on
ditio
ns
0 20 40 60 800%
50%
100%
150%
200%
250%
nu
mb
er
of
sky c
on
ditio
ns
pe
rce
nt
of
sky c
on
ditio
ns
geometrical CF from MYD0 20 40 60 80
0
100
200
300
400
500
600
700
(e)
0 20 40 60 800%
50%
100%
150%
200%
250%
nu
mb
ers
of
sky c
on
ditio
ns
pe
rce
nt
of
sky c
on
ditio
ns
0 20 40 60 800
100
200
300
400
500
600
700
(f)
geometrical CF from MOD
0 20 40 600%
50%
100%
150%
200%
250%
nu
mb
er o
f sk
y c
on
dit
ion
s
0 20 40 60 800
100
200
300
400
500(g)
per
cen
t o
f sk
y c
on
dit
ion
s
COT from MYD0 20 40 60
0%
100%
200%
300%
400%
num
ber
s of
sky c
ondit
ions
per
cent
of
sky c
ondit
ions
COT from MOD0 20 40 60 80
0
100
200
300
400
500(h)
VCD from geometrical method of MAX-
DOAS measurements
VCD from profile retrieval process of
MAX-DOAS measurements
statement R2 slope Point number R2 slope Point number
All points 0.6585 1.3717 203 0.578 1.213 142
CF
